TC Online First, published on March 21, 2018 as 10.1136/tobaccocontrol-2017-054117 Research paper Tob Control: first published as 10.1136/tobaccocontrol-2017-054117 on 21 March 2018. Downloaded from Changes in smoking cessation assistance in the European Union between 2012 and 2017: pharmacotherapy versus counselling versus e-cigarettes Filippos T Filippidis,1,2 Anthony A Laverty,1 Ute Mons,3 Carlos Jimenez-Ruiz,4 Constantine I Vardavas2,5,6

1Public Health Policy Evaluation Abstract and the WHO Framework Convention for Tobacco Unit, School of Public Health, Background The landscape of smoking cessation may Control (FCTC).2 Imperial College London, There is, however, wide variation between EU London, UK have changed in Europe recently. 2Center for Health Services Objectives To identify changes in use of smoking Member States (EU MS) in both their smoking prev- Research, School of Medicine, cessation assistance in the European Union (EU) and alence and in the extent of their tobacco control National and Kapodistrian factors associated with use of cessation assistance. policies. A 2014 review identified gaps in the imple- University of Athens, Athens, Methods Data from the 2012 (n=9921) and mentation of the FCTC in Europe, including high- Greece 3Cancer Prevention Unit and 2017 (n=9489) waves of the Eurobarometer survey lighting concerns that the region was progressing WHO Collaborating Centre were used. Self-reported use of smoking cessation slowly in terms of providing comprehensive cessa- for Tobacco Control, German assistance was assessed among smokers who had ever tion services—an important aspect of FCTC Article Cancer Research Center, tried to quit and former smokers. Changes in use of 14.3 A previous analysis of Eurobarometer data Heidelberg, Germany 4 across the EU found that the majority of smokers Smoking Cessation Service, each type of assistance were assessed using logistic Community of Madrid, Madrid, regression. who were trying to quit tobacco use were not using Spain Results Among current and former smokers, those who assistance. Moreover, the use of cessation assistance 5Institute of Public Health, had ever attempted to quit without assistance increased varies among countries, depending in part on their American College of Greece, from 70.3% (2012) to 74.8% (2017). Current smokers policies regarding access to cessation methods.4 This Athens, Greece 6Nicotine Dependence Center, were more likely to have used any assistance compared variation in provision of cessation assistance has copyright. Mayo Clinic, Rochester, with former smokers (P<0.001). Use of e-cigarettes raised concerns over how people are attempting to Minnesota, USA for smoking cessation assistance increased (3.7% to quit, especially if they are unable to access evidence- 9.7%)%), while use of pharmacotherapy (14.6% to based smoking cessation methods. This is particu- Correspondence to 11.1%)%) and smoking cessation services (7.5% to larly important for clinicians dealing with patients Dr Filippos T Filippidis, 5.0%)%) declined. Younger people were more likely to trying to quit with a variety of methods for which Department of Primary Care and Public Health, School of have reported e-cigarette use for smoking cessation but they may have little experience or limited evidence.

Public Health, Imperial College less likely to have used a cessation service. Individuals In light of this gap, we conducted secondary anal- http://tobaccocontrol.bmj.com/ London, London W6 8RP, UK; living in countries with comprehensive smoking cessation yses of pooled Eurobarometer datasets from 2012 f​ .​filippidis@imperial​ .​ac.uk​ policies were more likely to have used any cessation and 2017 in order to assess the changing landscape of smoking cessation assistance used across EU MS Received 25 October 2017 assistance (adjusted OR (aOR)=1.78; 95% CI 1.15 Revised 25 February 2018 to 2.76), pharmacotherapy (aOR=3.44; 95% CI 1.78 during the past 5 years among current and former Accepted 26 February 2018 to 6.66) and smoking cessation services (aOR=2.27; smokers. As a secondary outcome, we also inves- 95% CI 1.27 to 4.06) compared with those living in tigated potential EU MS-specific changes in the countries with weak smoking cessation policies. use of cessation assistance and sociodemographic Conclusions These findings highlight the need for factors associated with the use of particular types approaches that ensure that smokers get support to of assistance. quit smoking across the EU. The question of whether the

availability of e-cigarettes will displace other methods, on September 29, 2021 by guest. Protected Methods and the impact of such a displacement, should be closely Data source evaluated. We analysed data from 27 EU MS, collected in two Eurobarometer surveys: wave 77.1 (February–March 2012) and wave 87.1 (March Introduction 2017).5 6 All Eurobarometer surveys employ a The WHO has selected a 30% reduction in tobacco multistage sampling design in each MS according use as one of the 25 by 2025 goals, and the WHO to which primary sampling units (PSUs) are selected Regional Office for Europe has professed their from each region within each country, propor- ultimate goal to have a European region free of tional to population size. Subsequently, a sample tobacco use.1 Smoking cessation is thus a core of starting addresses is randomly selected in each To cite: Filippidis FT, activity for population health and entails substan- PSU, and households are systematically selected Laverty AA, Mons U, et al. Tob Control Epub ahead of tial health benefits both for individual smokers and following a standard random route starting from print: [please include Day public health. Smoking cessation has additionally these initial addresses. One participant in each Month Year]. doi:10.1136/ been recognised as an essential component of the household is randomly selected and interviewed tobaccocontrol-2017-054117 WHO’s MPOWER package for tobacco control in the local language. The methodology has been

Filippidis FT, et al. Tob Control 2018;0:1–6. doi:10.1136/tobaccocontrol-2017-054117 1 Copyright Article author (or their employer) 2018. Produced by BMJ Publishing Group Ltd under licence. Research paper Tob Control: first published as 10.1136/tobaccocontrol-2017-054117 on 21 March 2018. Downloaded from consistent across Eurobarometer waves. The European Commis- country that could be potentially classified into two categories (7 sion does not publish response rates; however, poststratification points in 2013 and 8 points in 2016). We used the most recent and population size weighting were applied in each country/ score to classify it as a high-scoring country. region to produce nationally representative samples in terms of age, gender and area of residence. A total of n=26 751 and Statistical analysis n=26 853 individuals aged ≥15 years from 27 EU MS were We restricted our analysis to former smokers and current interviewed in 2012 and 2017, respectively. A total of 19 410 smokers who reported having tried to quit in the past. Descrip- respondents (9921 in 2012 and 9489 in 2017) were included in tive results are presented as % with 95% CIs and regression the analysis as either former smokers or current smokers who results as adjusted ORs with 95% CI. Proportions are compared reported a previous quit attempt. Interviews in Croatia, the 28th with χ2 tests. Multilevel logistic regression models with country EU MS, were only conducted in 2017; therefore, it was excluded as the higher level of analysis were fitted to assess the association from this analysis. between independent variables (age, sex, education, difficulty paying bills, area of residence, TCS treatment score and year of Measures survey) and each of the following outcomes: (A) having used any Tobacco smoking cessation assistance; (B) having tried to quit without assistance; All respondents were asked ‘Regarding smoking cigarettes, cigars (C) having used pharmacotherapy; (D) having used smoking or a pipe, which of the following applies to you?’. Responses cessation services; and (E) having used electronic cigarettes or included ‘You currently smoke’ (ie, current smokers); ‘You used similar devices. The multilevel model controls for clustering to smoke but you have stopped’ (ie, former smokers); and ‘You within countries. An interaction term between TCS treatment have never smoked’ (ie, never smokers). score and calendar year was included to explore whether change in use of cessation assistance differed according to comprehen- Use of cessation assistance siveness of smoking cessation policies. Associations with the Former smokers, as well as current smokers who had ever tried remaining cessation types of assistance were not assessed, as the to quit smoking in the past were asked: ‘Which of the following number of responses was too low to be able to draw any reliable did you use in order to quit or to try to quit smoking?’. Respon- results. Separate models, adjusting for the same variables, were dents could choose one or more responses. Wording of some run for each EU MS to assess differences between 2012 and 2017 responses differed slightly between the two surveys, but for the in the reported use of the three most commonly reported types purpose of this analysis, we grouped them as follows: (A) medica- of assistance: pharmacotherapy, smoking cessation services and tion, including nicotine replacement medications (nicotine gum, electronic cigarettes. Finally, correlation coefficients between patch, inhaler and so on (nicotine replacement therapy (NRT))) country level changes (ie, change in ever use between 2012 and copyright. or other medications (’Pharmacotherapy’); (B) health services, 2017 as a percentage of the 2012 value) in ever use of phar- including support from the doctor or other health professional macotherapy, smoking cessation services and e-cigarettes were or special stop-smoking services (clinics, specialists, quitlines and calculated. All analyses were performed with Stata V.14.0, and so on) (‘Smoking cessation services’); (C) oral tobacco (snus), weights provided in the Eurobarometer dataset were used in chewing or nasal tobacco (snuff) (‘Oral or nasal tobacco’); (D) descriptive analyses to account for the complex sampling design electronic cigarettes or similar devices (‘Electronic cigarettes’); of the survey.

(E) other; and (f) you quit or you tried to quit without assistance http://tobaccocontrol.bmj.com/ (‘Without assistance’). Respondents who mentioned at least one Results of options (A) to (E) were classified as having used any smoking Among current and former smokers, the majority of respondents cessation assistance in the past. had tried to quit without assistance both in 2012 (70.3%) and in 2017 (74.8%) (table 1). Pharmacotherapy was the most popular Sociodemographic data cessation assistance (14.6% in 2012 and 11.1% in 2017), Respondents also provided data on their age (15–24, 25–39, followed by electronic cigarettes and cessation services. More 40–54 and ≥55 years), sex (male; female), age at which they current smokers who had attempted to quit reported having stopped full-time education (≤15, 16–19 and ≥20 years old), used any cessation assistance compared with former smokers their difficulties to pay bills during the last 12 months (almost both in 2012 (40.4% vs 25.6%, P<0.001) and in 2017 (37.5% never/never and from time to time/most of the time) and area of vs 20.8%, P<0.001). Among those who had ever used any assis- on September 29, 2021 by guest. Protected residence (rural and urban). tance to quit smoking, 16.6% of respondents in 2012 and 12.1% in 2017 (P<0.001) reported having ever used more than one National cessation policies type of assistance. Including ‘without assistance’ as a cessation We collected data on national smoking cessation policies from method, 9.5% of our sample in 2012 and 6.0% of our sample in the Tobacco Control Scale (TCS) 20137 and 2016,8 which 2017 reported having used multiple methods to quit (P<0.001). score European countries according to their tobacco control In 2017, 35.3% of those who had ever used pharmacotherapy, policies. We used the score for the subscale ‘treatment to help 34.9% of those who had ever used cessation services, 32.1% smokers stop’, which can range from 1 to 10 points and evalu- of those who had ever used e-cigarettes to quit and only 4.1% ates recording of smoking status in medical notes (1 point); brief of those who had ever tried without assistance reported having advice in primary care (1 point); quitline (2 points); network of tried to quit with multiple methods. smoking cessation support and its reimbursement (4 points); and Former smokers and current smokers who had tried to quit reimbursement of medications (2 points). We classified member were less likely to have used any cessation assistance in 2017, states into three categories: low (1–4), medium (5–7) and high compared with 2012 (aOR=0.61; 95% CI 0.52 to 0.71) and TCS treatment score (8–10). Country-specific scores varied very consequently more likely to have tried quitting without any assis- little between 2013 and 2016; therefore, the classification for tance (aOR=1.72; 95% CI 1.47 to 2.02), but these changes were each member state was consistent over time. Ireland was the only attenuated in countries with medium and high TCS treatment

2 Filippidis FT, et al. Tob Control 2018;0:1–6. doi:10.1136/tobaccocontrol-2017-054117 Research paper Tob Control: first published as 10.1136/tobaccocontrol-2017-054117 on 21 March 2018. Downloaded from

Table 1 Use of smoking cessation assistance among former smokers and current smokers who have tried to quit in the European Union, 2012– 2017 Current smokers who had ever tried to quit Former smokers Total % (95% CI) % (95% CI) % (95% CI) 2012 2017 2012 2017 2012 2017 n=4341 n=3489 n=5580 n=6000 n=9921 n=9489 Pharmacotherapy 22.5 (20.5 to 24.5) 16.6 (14.6 to 18.8) 8.5 (7.5 to 9.7) 7.5 (6.5 to 8.7) 14.6 (13.6 to 15.8) 11.1 (10.1 to 12.3) Smoking cessation 8.5 (7.3 to 9.9) 5.7 (4.5 to 7.1) 6.7 (5.8 to 7.8) 4.5 (3.7 to 5.4) 7.5 (6.7 to 8.4) 5.0 (4.3 to 5.8) services Electronic cigarettes 7.1 (5.9 to 8.4) 15.6 (13.5 to 17.8) 1.1 (0.7 to 1.5) 5.8 (4.8 to 6.9) 3.7 (3.2 to 4.3) 9.7 (8.7 to 10.8) Oral or nasal tobacco 0.6 (0.4 to 0.9) 1.1 (0.7 to 1.9) 0.7 (0.6 to 1.1) 1.2 (0.9 to 1.6) 0.7 (0.5 to 0.9) 1.2 (0.8 to 1.5) Without assistance 65.7 (63.4 to 67.9) 66.2 (63.4 to 68.8) 73.9 (72.1 to 75.7) 80.7 (78.9 to 82.3) 70.3 (68.9 to 71.7) 74.8 (73.3 to 76.3) Any assistance 40.4 (38.1 to 42.7) 37.5 (34.8 to 40.3) 25.6 (23.8 to 27.4) 20.8 (19.2 to 22.6) 32.2 (30.8 to 33.7) 27.5 (26.0 to 29.1) scores as shown by the interaction terms of TCS treatment score (aOR=0.53; 95% CI 0.38 to 0.74) in 2017 when compared and year (table 2). The reported use of e-cigarettes as cessation with 2012. The effect size for use of health professional/cessa- assistance was almost two times higher in countries with low tion services did not differ among countries with different TCS TCS treatment score in the 2017 wave (aOR=1.85; 95% CI 1.31 treatment scores. to 2.61) and much higher in countries with medium and high People living in countries with more comprehensive smoking TCS treatment scores. In contrast, respondents were less likely cessation policies (ie, with high TCS treatment scores) were more to report having used pharmacotherapy (aOR=0.73; 95% CI likely to have used any cessation assistance (aOR=1.78; 95% CI 0.58 to 0.93) and health professional/cessation service advice 1.15 to 2.76), pharmacotherapy (NRT or other medication)

Table 2 Associations of self-reported smoking cessation assistance with sociodemographic factors and tobacco treatment policies in 27 EU MS among former smokers and current smokers who had tried to quit, 2012–2017 (n=19 410) Any assistance* aOR Without assistance aOR Pharmacotherapy aOR Health services aOR Electronic cigarettes aOR

(95% CI) (95% CI) (95% CI) (95% CI) (95% CI) copyright. Year 2012 (ref) 1.00 1.00 1.00 1.00 1.00 2017 0.61 (0.52 to 0.71) 1.72 (1.47 to 2.02) 0.73 (0.58 to 0.93) 0.53 (0.38 to 0.74) 1.85 (1.31 to 2.61) Tobacco Control Scale treatment score Low (ref) 1.00 1.00 1.00 1.00 1.00 Medium 1.20 (0.86 to 1.68) 0.92 (0.68 to 1.26) 1.59 (0.95 to 2.65) 1.47 (0.93 to 2.31) 1.19 (0.72 to 1.97) High 1.78 (1.15 to 2.76) 0.57 (0.38 to 0.86) 3.44 (1.78 to 6.66) 2.27 (1.27 to 4.06) 1.58 (0.83 to 3.01) http://tobaccocontrol.bmj.com/ Year*Tobacco Control Scale treatment score 2017*Low (ref) 1.00 1.00 1.00 1.00 1.00 2017*Medium 1.39 (1.17 to 1.66) 0.64 (0.54 to 0.77) 1.17 (0.90 to 1.51) 1.22 (0.85 to 1.77) 1.28 (0.87 to 1.89) 2017*High 1.43 (1.14 to 1.78) 0.64 (0.51 to 0.80) 0.69 (0.50 to 0.93) 0.81 (0.51 to 1.28) 2.42 (1.53 to 3.82) Age (years) ≥55 (ref.) 1.00 1.00 1.00 1.00 1.00 40–54 1.45 (1.34 to 1.58) 0.73 (0.67 to 0.79) 1.69 (1.52 to 1.88) 0.91 (0.78 to 1.06) 2.32 (1.94 to 2.77) 25–39 1.32 (1.21 to 1.45) 0.81 (0.73 to 0.88) 1.33 (1.18 to 1.50) 0.66 (0.55 to 0.79) 2.89 (2.40 to 3.47) 15–24 1.38 (1.20 to 1.60) 0.80 (0.69 to 0.92) 0.88 (0.71 to 1.10) 0.54 (0.38 to 0.74) 4.38 (3.43 to 5.60)

Sex on September 29, 2021 by guest. Protected F emale (ref.) 1.00 1.00 1.00 1.00 1.00 Male 0.98 (0.92 to 1.05) 1.03 (0.96 to 1.10) 0.89 (0.82 to 0.98) 1.00 (0.88 to 1.13) 1.06 (0.93 to 1.21) Difficulties paying bills Never/almost never (ref.) 1.00 1.00 1.00 1.00 1.00 From time to time/most of 1.19 (1.10 to 1.28) 0.84 (0.78 to 0.91) 1.22 (1.11 to 1.35) 1.29 (1.12 to 1.49) 1.35 (1.17 to 1.57) the time Age when stopped education (years) Up to 15 (ref.) 1.00 1.00 1.00 1.00 1.00 16–19 1.18 (1.06 to 1.31) 0.88 (0.79 to 0.98) 1.33 (1.15 to 1.54) 1.21 (1.00 to 1.46) 1.30 (1.03 to 1.63) ≥20 1.12 (1.01 to 1.25) 0.93 (0.83 to 1.04) 1.24 (1.06 to 1.44) 1.26 (1.03 to 1.55) 1.15 (0.90 to 1.46) Area of residence Rural (ref.) 1.00 1.00 1.00 1.00 1.00 Urban 1.07 (1.00 to 1.15) 0.94 (0.88 to 1.01) 1.04 (0.95 to 1.15) 1.12 (0.98 to 1.28) 1.13 (0.97 to 1.31) *Includes pharmacotherapy, smoking cessation services, electronic cigarettes, oral or nasal tobacco and other. aOR, adjusted OR; EU MS, EU Member States.

Filippidis FT, et al. Tob Control 2018;0:1–6. doi:10.1136/tobaccocontrol-2017-054117 3 Research paper Tob Control: first published as 10.1136/tobaccocontrol-2017-054117 on 21 March 2018. Downloaded from

Figure 1 Changes (aOR with 95% CI) between 2012 and 2017 in having used pharmacotherapy, cessation services and e-cigarettes, respectively, as cessation assistance among former smokers and current smokers who had tried to quit in 27 EU member states (n=19 410). *Cessation services include support from the doctor or other health professional or special smoking cessation services (clinics, specialists, quitlines and so on). Pharmacotherapy includes nicotine replacement medications (nicotine gum, patch, inhaler and so on) or other medications. aOR, adjusted OR; EU, European Union.

(aOR=3.44; 95% CI 1.78 to 6.66) and health professionals/ differing depending on the level of tobacco treatment policies. smoking cessation services (aOR=2.27; 95% CI 1.27 to 4.06) Notably, in countries with more comprehensive smoking cessa- compared with those living in countries with low TCS treatment tion policies, adults reported substantially higher use of phar- scores. There was no statistically significant difference regarding macotherapy and cessation services, indicating the impact of the use of e-cigarettes. national policies on cessation attempts. copyright. Younger people (15–24 years) were more likely than those The increase in the reported use of electronic cigarettes as aged ≥55 years to have used any cessation assistance including cessation assistance coinciding with a decrease in the use of phar- e-cigarettes (aOR 4.38; 95% CI 3.43 to 5.60) but less likely to macotherapy or support from health professionals is consistent have received advice from health professionals/smoking cessa- with findings from the English Smoking Toolkit study conducted tion services (aOR 0.54; 95% CI 0.38 to 0.74). People with diffi- between 2011 and late 2014.9 That study concluded that there culties paying bills were more likely to have used any cessation was no evidence that the decline in the use of NRT could be assistance than those who never or almost never have problems attributed to the rise in the use of e-cigarettes, a finding they paying their bills. Sex, with the exception of pharmacotherapy, confirmed also using more recent data.10 Use of medication and http://tobaccocontrol.bmj.com/ and area of residence were not significantly associated with the NRT for smoking cessation can be influenced by factors such as use of cessation assistance. the availability of new pharmaceuticals, funding and smoke-free Variations in changes between 2012 and 2017 across EU MS legislation11–14; therefore, the explanation of these changes may are shown in figure 1. The reported past use of pharmacotherapy differ between EU MS. In our study, it is not clear whether using was less likely in eight of the 27 EU MS in 2017, compared with e-cigarettes as a substitute for tobacco impacts use of established 2012 and more likely only in Italy, while the remaining 18 MS services like NRT and smoking cessation clinics or whether it experienced non-statistically significant changes. A statistically is mostly employed by smokers who would otherwise try to significant increase in the use of electronic cigarettes to quit quit without assistance, especially due to the inexistence of a smoking was reported in 12 EU MS. In nine EU MS, support from supportive policy environment for standard cessation assistance health professionals/smoking cessation services was less likely to (pharmacotherapy and cessation services). At the individual MS be reported in 2017, compared with 2012, with no statistically level, there was only weak correlation between changes in e-cig- on September 29, 2021 by guest. Protected significant increase in any of the EU MS (figure 1). Changes in arette use and changes in use of standard cessation assistance, pharmacotherapy and cessation services use at the MS level were which might suggest coincidence rather than displacement. modestly correlated (r=0.53). On the contrary, change in e-ciga- More research on whether e-cigarette use is displacing standard rette use was weakly correlated with both changes in pharmaco- cessation assistance in Europe and how this may impact long- therapy (r=0.22) and in cessation services (r=0.18). term abstinence is required, especially considering the increase in e-cigarette use.15 16 Discussion While changes in cessation assistance and support used were This secondary analysis of two recent Eurobarometer waves quite similar in the majority of EU MS, we nevertheless observed found that the use of any self-reported smoking cessation assis- some heterogeneity between countries. As a previous interna- tance decreased in the EU between 2012 and 2017. Self-reported tional study of differences in cessation assistance in 15 countries use of e-cigarettes for smoking cessation increased, while the use noted, such variation may reflect a combination of differences of standard cessation assistance, such as pharmacotherapy and in tobacco control efforts and priorities, the affordability of support from healthcare professionals and cessation services, different methods and more general cultural factors.17 In our was less likely to be reported in 2017 compared with 2012. analyses, national policies regarding availability and cost of There was considerable variation between EU MS, with changes cessation services were associated with both use of the assessed

4 Filippidis FT, et al. Tob Control 2018;0:1–6. doi:10.1136/tobaccocontrol-2017-054117 Research paper Tob Control: first published as 10.1136/tobaccocontrol-2017-054117 on 21 March 2018. Downloaded from cessation assistance and changes over time. Our findings are in reported much lower use of cessation assistance compared with line with an earlier analysis of Eurobarometer data from 2012, smokers who had tried to quit without success. This is likely which found that living in a country that offers cost-covered explained by the fact that smokers who find it more difficult to national quitlines, medication and other cessation services was quit are also more likely to seek assistance; therefore, our study associated with higher likelihood of using standard cessation cannot provide insight into the relative effectiveness of smoking assistance with proven efficacy.4 Although there seems to be an cessation methods. overall decline in use of cessation assistance for smoking cessa- tion across the EU, this concerning trend was attenuated in MS Strengths and limitations with more comprehensive cessation services, highlighting the We used two cross-sectional surveys with large samples repre- importance of national policies and FCTC Article 14 imple- sentative of the EU population and consistent sampling meth- mentation. It has to be noted, however, that the question used odology; therefore, we are confident that these conclusions are for the analysis refers to ‘ever use of cessation assistance’, so it generalisable to the entire EU population, although the samples would take a considerable amount of time to detect an effect of at the country level were modest, typically around 1000 indi- any policy change, as there is already a substantial number of viduals in each member state. The samples were independently people who had used certain cessation assistance in quit attempts selected in the two waves (no cohort was followed up), hence before the current policies were put into place. This limitation any causal interpretations of the detected associations should of our analysis further highlights the increase in e-cigarette use be made with caution. Further cohort study data across the EU for cessation, as it is a product that has become available much is needed.26 The classification of policies regarding cessation more recently compared with other types of cessation assistance, services may also not provide adequate granularity to capture although respondents may have been more likely to recall assis- small but potentially important variations in policies between EU tance used in more recent quit attempts. MS. Additionally, we had no data regarding the time when each Our results have important implications for clinical practice, type of cessation assistance had been used, which would have as they highlight the fact that European healthcare professionals helped us identify changes specifically associated with recent are increasingly having to deal with smokers who are using e-cig- use of cessation assistance, nor was any information on efficacy arettes for smoking cessation and are less likely to use well estab- available. All data were self-reported and thus may be subject to 18–21 lished cessation assistance. Regardless of factors related to bias, as people may be more likely to remember cessation assis- access and cost of cessation services, our findings may also reflect tance used more recently or, in the case of former smokers, those the reluctance of some smokers to engage in ‘medicalised’ cessa- that were effective in helping them to quit. Former smokers were tion that involves contact with health services or use of medica- defined as such based on self-reports; no minimum period of tion. Beyond questions surrounding the efficacy of e-cigarettes abstinence was reported. Finally, we combined data for ciga- copyright. 22 in smoking cessation, for which there is limited evidence from rettes, cigars and pipes; however, because regular use of cigars 23 randomised controlled trials, this trend poses a new challenge and pipes is rare in the EU,27 any effect on our results is most for health professionals, whose experience in handling smoking likely minimal. cessation among smokers who use e-cigarettes or wish to use e-cigarettes is still limited. This challenge also extends to regu- Conclusion lating bodies, which should closely monitor research on novel The majority of attempts to quit smoking in the EU continue to products that are or may be used in quit attempts and consider

be without any cessation assistance. The proportion of smokers http://tobaccocontrol.bmj.com/ updating their guidance to clinicians accordingly. attempting to quit using assistance has further decreased over It is also important to explore country-specific factors that the past 5 years. The use of established aids such as pharmaco- may have influenced the use of smoking cessation assistance in therapy have become less popular, while e-cigarettes as a poten- recent years and could explain the variability in changes between tial cessation or switching method has grown between 2012 and member states. In the UK, for example, increase in e-cigarette 2017. This highlights the need to evaluate their efficacy and use was coupled with a decline in use of pharmacotherapy and impact on individual abstinence as well as their population-level cessation services, although these opposing trends may not be implications and to explore whether they may be displacing stan- 9 10 14 associated. On the contrary, despite an increase in e-cig- dard cessation assistance. Finally, it is important to emphasise the arette use for smoking cessation in the Netherlands, use of central role of the smoking cessation policy environment. Living pharmacotherapy and cessation services has not changed, while other member states, such as Germany and Poland, experienced on September 29, 2021 by guest. Protected decreased use of pharmacotherapy and cessation services without What this paper adds a concurrent change in the use of e-cigarettes over the study period. It is also interesting to note that no statistically signifi- ►► Use of e-cigarettes for smoking cessation has increased, cant changes were reported in Greece, despite worsening access while use of pharmacotherapy and smoking cessation to healthcare services overall during the economic crisis.24 These services has decreased across the European Union between examples underscore the complexity of the smoking cessation 2012 and 2017. landscape in the EU and the multitude of factors that may shape ►► Changes differed widely among member states. Individuals future developments in each member state. living in countries with comprehensive smoking cessation In our analysis, we found that a number of smokers have tried policies were more likely to have used any cessation to quit smoking using multiple methods, including a variety of assistance. cessation assistance and trying without assistance. This is consis- ►► Smoking cessation is a key strategy in the battle against tent with findings from the USA,25 although the difference in tobacco both at a population and at the individual levels. the cessation assistance assessed does not allow direct compar- Identifying trends and factors associated with the use of isons. Moreover, our data did not allow us to explore whether cessation assistance can inform both policy makers and different types of cessation assistance were used concurrently or clinicians. in distinct quit attempts. In addition, those who successfully quit

Filippidis FT, et al. Tob Control 2018;0:1–6. doi:10.1136/tobaccocontrol-2017-054117 5 Research paper Tob Control: first published as 10.1136/tobaccocontrol-2017-054117 on 21 March 2018. Downloaded from in a country that offers cost-covered national quitlines, phar- 6 European Commission. Eurobarometer 87.1, March 2017. GESIS Data Archive: macotherapy and other cessation services was associated with ZA6861, Data file Version 1.2.0,10.4232/1.12922. Brussels: TNS OPINION & SOCIAL, 2017. higher likelihood of using standard cessation assistance with 7 Joossens L, Raw M. The tobacco control scale 2013 in Europe, 2013. proven efficacy, highlighting the importance of national policies 8 Joossens L, Raw M. The tobacco control scale 2016 in Europe, 2017. in FCTC Article 14 implementation. 9 Beard E, Brown J, McNeill A, et al. Has growth in electronic cigarette use by smokers been responsible for the decline in use of licensed nicotine products? Findings from repeated cross-sectional surveys. Thorax 2015;70:974–8. Acknowledgements The Public Health Policy Evaluation Unit at Imperial College 10 Beard E, West R, Michie S, et al. Association between electronic cigarette use is grateful for the support of the NIHR School of Public Health Research. and changes in quit attempts, success of quit attempts, use of smoking cessation Contributors FTF had the main role in study conception and data analysis. All pharmacotherapy, and use of stop smoking services in England: time series analysis of authors contributed to manuscript preparation and data interpretation. All authors population trends. BMJ 2016;354:i4645. have read and approved the final version of the manuscript. 11 Szatkowski L, Coleman T, McNeill A, et al. The impact of the introduction of smoke- free legislation on prescribing of stop-smoking medications in England. Addiction Funding This work was supported by a grant from the European Commission 2011;106:1827–34. (Horizon2020 HCO-6- 2015; EUREST-PLUS: 681109; Vardavas). AAL receives funding 12 Jarlenski M, Hyon Baik S, Zhang Y. Trends in use of medications for smoking cessation from the NIHR (RP 2014- 04-032). in medicare, 2007-2012. Am J Prev Med 2016;51:301–8. Competing interests None declared. 13 Kasza KA, Cummings KM, Carpenter MJ, et al. Use of stop-smoking medications in the United States before and after the introduction of varenicline. Addiction Patient consent Detail has been removed from this case description/these case 2015;110:346–55. descriptions to ensure anonymity. The editors and reviewers have seen the detailed 14 Britton J. Electronic cigarettes and smoking cessation in England. BMJ information available and are satisfied that the information backs up the case the 2016;354:i4819. authors are making. 15 Filippidis FT, Laverty AA, Gerovasili V, et al. Two-year trends and predictors of Ethics approval Data were freely available and deidentified, thus no ethical e-cigarette use in 27 European Union member states. Tob Control 2017;26. approval was required. 16 Filippidis FT, Laverty AA, Vardavas CI. Experimentation with e-cigarettes as a smoking Provenance and peer review Not commissioned; externally peer reviewed. cessation aid: a cross-sectional study in 28 European Union member states. BMJ Open 2016;6:e012084. Data sharing statement The datasets supporting the conclusions of this article 17 Borland R, Li L, Driezen P, et al. Cessation assistance reported by smokers in 15 are freely available online, doi:10.4232/1.12265 in www.​gesis.​org. countries participating in the International Tobacco Control (ITC) policy evaluation Open Access This is an Open Access article distributed in accordance with the surveys. Addiction 2012;107:197–205. Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which 18 Cahill K, Stevens S, Perera R, et al. Pharmacological interventions for smoking permits others to distribute, remix, adapt, build upon this work non-commercially, cessation: an overview and network meta-analysis. Cochrane Database Syst Rev and license their derivative works on different terms, provided the original work 2013;5:CD009329. is properly cited and the use is non-commercial. See: http://​creativecommons.​org/​ 19 Stead LF, Perera R, Bullen C, et al. Nicotine replacement therapy for smoking licenses/by-​ ​nc/4.​ ​0/ cessation. Cochrane Database Syst Rev 2012;11:CD000146. 20 Stead LF, Buitrago D, Preciado N, et al. Physician advice for smoking cessation.

© Article author(s) (or their employer(s) unless otherwise stated in the text of the Cochrane Database Syst Rev 2013;5:CD000165. copyright. article) 2018. All rights reserved. No commercial use is permitted unless otherwise 21 West R, Raw M, McNeill A, et al. Health-care interventions to promote and assist expressly granted. tobacco cessation: a review of efficacy, effectiveness and affordability for use in national guideline development. Addiction 2015;110:1388–403. 22 Kalkhoran S, Glantz SA. E-cigarettes and smoking cessation in real-world and clinical References settings: a systematic review and meta-analysis. Lancet Respir Med 2016;4:116–28. 1 World Health Organization - Regional Office for Europe. Roadmap of actions to 23 Hartmann-Boyce J, McRobbie H, Bullen C, et al. Electronic cigarettes for smoking strengthen implementation of the WHO Framework Convention on Tobacco Control in cessation. Cochrane Database Syst Rev 2016;9:CD010216. the European Region 2015–2025: making tobacco a thing of the past, 2015. 24 Filippidis FT, Gerovasili V, Millett C, et al. Medium-term impact of the economic crisis 2 World Health Organization. WHO report on the global tobacco epidemic, 2008: The on mortality, health-related behaviours and access to healthcare in Greece. Sci Rep http://tobaccocontrol.bmj.com/ MPOWER package. Geneva, 2008. 2017;7:46423. 3 World Health Organization. European tobacco control status report 2014: WHO - 25 Caraballo RS, Shafer PR, Patel D, et al. Quit methods used by us adult cigarette Regional Office for Europe, 2014. smokers, 2014-2016. Prev Chronic Dis 2017;14:E32. 4 Filippidis FT, Gerovasili V, Vardavas CI, et al. Determinants of use of smoking cessation 26 Vardavas C. EUREST PLUS - european regulatory science on tobacco: policy aids in 27 European countries. Prev Med 2014;65:99–102. implementation to reduce lung diseases - proposal (Horizon2020). Tob Prev Cessat 5 European Commission. Eurobarometer 77.1, February-March 2012. GESIS Data 2016;2. Archive: ZA5597, dataset version 2.0.0 (2012). Brussels: TNS OPINION & SOCIAL, 27 European Commission. Special Eurobarometer 458 “Attitudes of Europeans towards 2012. tobacco and electronic cigarettes”. Brussels, 2017. on September 29, 2021 by guest. Protected

6 Filippidis FT, et al. Tob Control 2018;0:1–6. doi:10.1136/tobaccocontrol-2017-054117